SAT ACT Scores Analysis
Abstract
There are many States with their declining SAT or ACT participation rates or worse, not even moved in between years. This paper proves to see what the trends are in the SAT and ACT scores and participation rates and digs deep using statistics to find out why some of it might be the case. At the end of it, it aims to have draw conclusions from the dataset, describes some limitations and finally recommends plausible solutions to the problem.
Summary of findings
Three states with interesting trends
From the percentage change between Class of 2017 and 2018 on the tests, Illinois and Colorado had the most positive trends and at the same time negative trends for the SAT and ACT tests respectively. Upon research both states went from mandatory ACT tests to SAT tests from the beginning of the 2016-2017 year [1][2] which was the transitional year for the switch and by the time when Class of 2018 was taking the SAT, it was already almost full participation.
Since the 2017-18 school year, all Rhode Island 11th graders must take the SAT. [3] This probably explains why there was an increase in SAT participation from 71% to 91% while there was a decrease in ACT participation year on year. Class of 2018 had not reached full participation yet probably because it was in its first transitional year.
Increasing accessibility and fairness
Based on the data exploration, our data show that with making a particular test mandatory or by offering free administration of the test, it increases the probability of the participation rate to be higher. The highest 6 States with a 100% participation rate on either years all are the result of making them mandatory or offered free.
The bottom two states with the lowest participation rates for SAT tests are Iowa and South Dakota.
Since they have the lowest rates we can start with one state which is Iowa. It seems to be the better choice with a much higher population [4] which in turn benefits more high school students to get a chance to go to University and a relatively median household income very close to the national median to support potential tuition fees. [5]
Image source: US Census Bureau [4]
Image source: Alaska Higher Education Almanac [5]
For more in depth analysis: you can refer to my Github for more information.